Diffing on undo steps is a critical performance point of override
system, although not required for override itself, it gives user
immediate feedback ove what is overridden.
Profiling showed that rna path text search over overrides operations was
by far the most costly thing here, so now using a runtime temp ghash
mapping for this search instead.
Seems to give at least 5 times speedup on big production rig.
Now, custom props defined as overriddable can be overridden, saved,
reloaded, etc. That fixes the last main issue with them.
Note that custom props still have a lot of glitches and weirdness in
their overriding behavior, but for now the most important is finally
achieved, will let them rest and settle a bit, those have been
incredibly painful to tame... :(
The triple possible status of an PropertyRNA (actual C-defined prop,
py-defined IDprop behaving similar ot RNA one, or actual 'pure' custom
prop) is really confusing... Not to mention the _RNA_UI ugly thingy on
top of that. :/
The old layout of `PointerRNA` was confusing for historic reasons:
```
typedef struct PointerRNA {
struct {
void *data;
} id;
struct StructRNA *type;
void *data;
} PointerRNA;
```
This patch updates it to:
```
typedef struct PointerRNA {
struct ID *owner_id;
struct StructRNA *type;
void *data;
} PointerRNA;
```
Throughout the code base `id.data` was replaced with `owner_id`.
Furthermore, many explicit pointer type casts were added which
were implicit before. Some type casts to `ID *` were removed.
Reviewers: brecht, campbellbarton
Differential Revision: https://developer.blender.org/D5558
That file was getting out of control, now comparison/override RNA code is
in `rna_access_compare_override.c`. 1K lines of code for now, but that
area is likely to grow more in the future...
Note that we can probably split more out of `rna_access.c`, but for now
that will do.
